The philosopher researcher
Within five years, the pharmaceutical industry will see about 40 per cent of its patents lose most of their value overnight. Meanwhile, drug discovery appears to be losing momentum and costs are rocketing, so any methodology which improves efficiency and lowers costs would not only save lives but money – in developing countries as well as in more affluent societies. …
Andrew Hopkins is not just building new computational and biophysical tools for designing new drugs but also questioning the underlying assumptions of the whole drug discovery process.
If you read his curriculum vitae, Professor Andrew Hopkins seems well qualified to work in a variety of scientific disciplines, including biophysics, chemistry and informatics, but even though he would love to be part of the team that comes up with the next new blockbuster drug, he is just as keen to talk about philosophy and poetry, and how they help us understand the mysteries of science. One moment, he is trying to explain “the global mapping of pharmacological space” and how to design “promiscuous drugs,” and the next he is discussing T. S. Eliot and Popper.
Whilst many scientists and pharmaceutical companies are desperate for original ideas, Hopkins is also concerned about where ideas come from. To illustrate the complexity of scientific creativity, he quotes from Eliot’s poem The Rock, and the “endless cycle of ideas and action” involved in invention, discussing the dramatic difference between “the eureka moment” and innovation – putting new ideas to practical use. “Science asks if it is true,” he explains, “while technology asks if it works.”
Hopkins is also concerned about “social technologies” or how to build the “virtuoso teams” which think up new concepts and develop new products, recognising that conventional structures and organisations do not always come up with the goods. Hopkins also thinks it is time to rethink the relationship between the pharmaceutical industry and academia, as we enter an era of greater openness in scientific research, when collaboration matters more than competition and researchers provide open access to data rather than treating it as an industrial secret.
Hopkins, who is SULSA Research Professor of Translational Biology and the Chair of Medicinal Informatics at the College of Life Sciences in the University of Dundee, also talks about the “paradox” of current pharmaceutical research, with more and more money (at least $80 billion a year) going into new projects, yet outcomes declining. The challenge, he says, is to translate the knowledge we already have into concrete results, including understanding how existing drugs work, and developing new technologies to improve productivity. “Drug discovery is still a cottage industry in many ways,” Hopkins explains. “There has not been the Darwinian pressure to evolve. Due to declining productivity, for the first time in four decades, the pharmaceutical industry is now facing a decline in revenues. And I believe this crisis of confidence is the ideal moment for innovation – there has never been a better time for academia to grasp the opportunities for new research.”
We also need a radical change in the way that new research is funded, says Hopkins, especially when it comes to neglected diseases, with non-governmental organisations and charities joining forces with academia and industry to drive major projects and scale up drug discovery for tropical diseases, for example.
“Tropical diseases offer academia and industry an invaluable opportunity to experiment with new ways of conducting drug discovery – organisationally, technologically and scientifically,” Hopkins continues. “We should be bold in proposing a grand ‘moon-shot’ type of mission in this area to develop ten new drugs for the ten major neglected diseases by 2020.”
Hopkins’ latest work in network pharmacology also suggests that the successful design of new drugs will need a major rethink of the way we think of targets, seeing them as part of complex networks rather than as individual targets which exist in isolation. “The dominant paradigm in drug discovery is the concept of designing maximally selective ligands to act on individual drug targets,” Hopkins wrote in Nature Chemical Biology last year. “However, many effective drugs act via modulation of multiple proteins rather than single targets.”
There may be few targets to aim for, he says, but the targets may have multiple receptors and the interactions between different targets and networks of targets make the process even more complex, and this is where network pharmacology may come to the rescue. “By targetting networks of proteins,” adds Hopkins, “we also circumvent toxicity.”
Part of Hopkins’ work in Dundee is to build the computational tools which will help to turn the concept of multiple targets into reality – mapping the networks of targets as well as the complex interactions between them. Hopkins also sees medicinal informatics “casting its nets” in research, not only modelling and analysing data from experiments but also exploring the existing published data to identify patterns which may lead to unforeseen discoveries. This is part of a process called chemogenomics – developing new methods to prioritise potential drug targets from a pathogen genome by analysing the collective pharmacology knowledge already available, including genome sequences, protein structures and literature abstracts.
Open access to data is another pet subject for Hopkins, who explains that we are “very dependent on data to develop medicinal informatics.” One of the first things he did when he came to Dundee was to work with others to persuade the Wellcome Trust to invest almost £5 million to create an open access database for medicinal chemistry data, to put more data into the public domain and “spur innovation.”
Computational tools, says Hopkins, could not only help to automate the drug discovery process, reducing costs and increasing efficiency, but also play a key role in clinical trials. Drug discovery accounts for about half of total research costs, but Hopkins sees the whole drug development process as a continuous cycle, with informatics feeding back data from clinical trials to sharpen available knowledge, thus making future trials more likely to succeed by understanding previous failures. In addition, says Hopkins, about 40 per cent of the drug sales on the market are the result of “alternative indications” – scientific accidents like Viagra, a drug which started off as a cardiovascular treatment rather than a cure for erectile dysfunction.